This table was derived from the data set (you mean "variable")Race. It shows that white people make up about 2/3 of the people surveyed. This first data set is important because it reflects that white people will have a large sway on the total survey. We can imply that the rest of the survey will reflect more white opinion compared to any other race. Education
No degree earned
460
Bachelor's degree
333
Master's degree
121
PhD, LIT, SCD, DFA, DLIT, DPH, DPHIL, JSC, SJD
16
LLB, JD
5
MD, DDS, DVM, MVSA, DSC, DO
8
JDC, STD, THD
2
Associate degree (AA)
260
NA's
1117
This table shows the levels of education completed by those who took the survey. The highest percent of those who answered the survey did completed any forms of higher education. Second to that are those who completed a Bachelor’s degree. Doctors and lawyers made a very small percentage of those surveyed. If you included the information from the Race table above, one could estimate that those surveyed are predominantly white people who work within the working and middle class. summary(nes$midclass) "Middle Class"
Min.
!st Qu.
Median
Mean
3rd Qu.
Max.
NA's
0
60
85
77.07
85
100
257
The summary of the Middle Class feeling thermometer shows a median of 85. This means that the average feeling towards the middle class is relatively high. The first 1st quartile proves this by showing the lower end has an average of 60 and the 3rd quartile shows an 85. The quartiles help show any reduce outliners that would throw off the mean. Standard deviation of Middle Class: 17.61054The average deviation from the mean was 17.6 is relatively low if you compare it to a scale of 1-100. This means that the answers were generally close to the mean and there was limited dispersion. summary(nes$workclass) "Working Class"
Min.
!st Qu.
Median
Mean
3rd Qu.
Max.
NA's
0
70
85
83.97
100
100
236
This summary shows that working class has a higher mean, 1st quartile, and 3rd quartile then middle class. The fact that the 3rd quartile and the max are the same indicates that the working class had several votes for maximum feeling. Standard deviation of Working Class: 16.52078The standard deviation of the working class was the smallest deviation among the four feeling thermometer categories I worked with. It shows that there were less outliners for working class compared to the other categories.
summary(nes$poorppl) "Poor People"
Min.
!st Qu.
Median
Mean
3rd Qu.
Max.
NA's
0
60
70
74.55
96.25
100
266
This table shows that the average feeling towards poor people was relatively high. However, the 1st Quartile compared to the 3rd quartile are very far apart. This shows that there were outliers within the survey.
The way you talk about outliers sounds as if outliers are the one and only manifestation of dispersion. Outliers definitely suggests a certain spread-out-ness in the data, but the standard deviation is a a more general measure of how spread out the data is. It says something about the whole distribution, whereas outliers only say there's extremity in one of the tails. Finally, distance between the 1st and 3rd quartile does not necessarily imply outliers, as outliers are observations far outside one of those quartiles!
Standard deviation for Poor People: 19.75
This standard of deviation is relatively high showing there are mixed feelings about poor people.
summary(nes$richppl) "Rich People"
Min.
!st Qu.
Median
Mean
3rd Qu.
Max.
NA's
0
50
50
58.28
70
100
279
This table shows that rich people were a little above average when it came to popularity. The 1st and 3rd quartile shows that there were more high outliers than there were lower.
Stand Deviation for Rich People: 21.40141
This is the highest deviation among the four categories i worked with. It shows that the mean does not completely represent the feelings for rich people. (Does it really mean that the mean is less representative of the distribution? Isn't it really just that the mean is always agnostic with respect to the variability?) There were a high precent of outliers for this category than the others listed.
The violin plots represent a cleaned up version of the information listed above. Violin plots are a combination of a box plot and a kernel density plot. Note the dot within the graph represents the median and the black box represents the quartile range. The width or "body" of the graph represent the density estimation. The orange violin plot representing rich people is very wide in the middle representing a high percentage of selections within that area. The red working class plot shows a more narrorw and spread selection. The plot quartile ranges all the way to the max 100. The plot that represents poor people shows a fairly consistent density from 60 to 100. This means the pole was fairly spread throughout that range. The middle class plot shows a considerable density within the 85-90 range. However, the mean for this graph was only 77.
Correlation between working class and middle class: 0.5841736 This number shows that there is a positive correlation between the working class and middle class.
Correlation between poor people and working class: 0.5102406 This number shows a positive correlation between poor people and the working class. It is common knowledge that those who work within the working class can be considered poor.
Correlation between rich people and middle class:
0.2904286
This number shows limited correlation between rich people and the middle class. This number is not large enough to confirm absolute correlation between the two variables.
Correlation between rich people and poor people:
.2809538
This number shows low positive correlation between rich people and poor people.
Scatter plots help compare two variables for a data set. In this scatter plot, feeling thermometer poles for middle class and working class were compared.This scatter plot first shows a slight negative correlation but shifts into a more dominate positive correlation. The correlation between working class and middle class was .584 which was the most dominate correlation I found.
This scatter plot compares rich people and poor people. If you recall, the correlation between the two was .28. This graph demonstrates a low positive correlation.
Great job, there's a lot here! You should keep thinking about how you might describe some of these measures, as I mentioned in comments above, but I'm just knitpicking...
Race
This table was derived from the data set (you mean "variable")Race. It shows that white people make up about 2/3 of the people surveyed. This first data set is important because it reflects that white people will have a large sway on the total survey. We can imply that the rest of the survey will reflect more white opinion compared to any other race.
Education
summary(nes$midclass) "Middle Class"
Standard deviation of Middle Class: 17.61054The average deviation from the mean was 17.6 is relatively low if you compare it to a scale of 1-100. This means that the answers were generally close to the mean and there was limited dispersion.
summary(nes$workclass) "Working Class"
Standard deviation of Working Class: 16.52078The standard deviation of the working class was the smallest deviation among the four feeling thermometer categories I worked with. It shows that there were less outliners for working class compared to the other categories.
summary(nes$poorppl) "Poor People"
The way you talk about outliers sounds as if outliers are the one and only manifestation of dispersion. Outliers definitely suggests a certain spread-out-ness in the data, but the standard deviation is a a more general measure of how spread out the data is. It says something about the whole distribution, whereas outliers only say there's extremity in one of the tails. Finally, distance between the 1st and 3rd quartile does not necessarily imply outliers, as outliers are observations far outside one of those quartiles!
Standard deviation for Poor People: 19.75
This standard of deviation is relatively high showing there are mixed feelings about poor people.
summary(nes$richppl) "Rich People"
Stand Deviation for Rich People: 21.40141
This is the highest deviation among the four categories i worked with. It shows that the mean does not completely represent the feelings for rich people. (Does it really mean that the mean is less representative of the distribution? Isn't it really just that the mean is always agnostic with respect to the variability?) There were a high precent of outliers for this category than the others listed.
The violin plots represent a cleaned up version of the information listed above. Violin plots are a combination of a box plot and a kernel density plot. Note the dot within the graph represents the median and the black box represents the quartile range. The width or "body" of the graph represent the density estimation. The orange violin plot representing rich people is very wide in the middle representing a high percentage of selections within that area. The red working class plot shows a more narrorw and spread selection. The plot quartile ranges all the way to the max 100. The plot that represents poor people shows a fairly consistent density from 60 to 100. This means the pole was fairly spread throughout that range. The middle class plot shows a considerable density within the 85-90 range. However, the mean for this graph was only 77.
Correlation between working class and middle class:
0.5841736
This number shows that there is a positive correlation between the working class and middle class.
Correlation between poor people and working class:
0.5102406
This number shows a positive correlation between poor people and the working class. It is common knowledge that those who work within the working class can be considered poor.
Correlation between rich people and middle class:
0.2904286
This number shows limited correlation between rich people and the middle class. This number is not large enough to confirm absolute correlation between the two variables.
Correlation between rich people and poor people:
.2809538
This number shows low positive correlation between rich people and poor people.
Scatter plots help compare two variables for a data set. In this scatter plot, feeling thermometer poles for middle class and working class were compared.This scatter plot first shows a slight negative correlation but shifts into a more dominate positive correlation. The correlation between working class and middle class was .584 which was the most dominate correlation I found.
This scatter plot compares rich people and poor people. If you recall, the correlation between the two was .28. This graph demonstrates a low positive correlation.
Great job, there's a lot here! You should keep thinking about how you might describe some of these measures, as I mentioned in comments above, but I'm just knitpicking...